webhosting   Cheap Reseller Hosting   links    free hosting by fateback   hosting reseller   100WebSpace offers 100MB Web Space

Writing the Arabic language with Tengwar characters

Introduction

This web-page discusses my ideas for a "mode" suitable for writing the Arabic language with J.R.R. Tolkien's Tengwar characters. A brief explanation by Tolkien of this alphabet can be found in Appendix E of his book The Lord of the Rings: The Return of the King, and there is also abundant information available on-line (see the links section below). On this page, the necessary diagrams appear as fixed-size (raster) images; go to the download section below to view a resizable (vector) PDF-file version of this web-page's content, or to download Tengwar Arabic fonts.

The Arabic Tengwar mode described here is a thought-experiment which shows how a number of annoying ambiguities, incompletenesses, and cases of lack of visual distinctiveness in the Arabic alphabet (as it is commonly used) might be remedied by using a non-cursive left-to-right alphabet, but while retaining a number of the basic structural features of current Arabic orthography -- and without resorting to Latin script (since the Latin alphabet is not suitable to write the Arabic language unless a large number of diacritic marks and/or new symbols were added, and since many Arabs would regard the use of the Latin alphabet to write Arabic as inappropriate cultural imperalism). In addition, the Tengwar Arabic mode has features which make it useful in transcribing not just standard written Arabic, but also Arabic colloquial dialects (something which is very difficult to do with the current Arabic-script orthography).

Current Arabic Orthography

Here is a very brief discussion of some of the difficulties with current Arabic orthography:

First, Arabic letters can take on radically different shapes depending on the surrounding context of the letters (if any) which occur preceding and following (see the horizontal row at the bottom of Table 1). And many of these contextual Arabic letter-forms are only distinguished from contextual forms of completely different Arabic letters (which represent completely different sounds) by means of detached diacritical dots which aren't part of the basic letter-shape at all (see the vertical column in Table 1). So a particular form of an Arabic letter (such as the medial form of ya') often resembles forms of other Arabic letters (such as the medial forms of the letters for "th", "t", "n", and "b") more than it does the other forms of the same letter (such as the initial, final, and isolated forms of ya') -- and often only the presence of the diacritical dots serves to distinguish letters which represent radically different sounds (such as [b] and [n]):

Table 1

Secondly, all the letters of the Arabic alphabet are originally basically consonantal, but some of them have come to be used to write long vowels. Therefore the letters which can write both consonant and long vowel sounds are significantly ambiguous -- unless the full panoply of Arabic-script diacritics is used (which it very seldom is, outside of Qur'ans, dictionaries, and children's textbooks). In the case of the letter 'alif (which is used to write both the glottal stop consonant sound and the long [a] vowel sound), the ambiguities are considered to be so severe that a whole complicated series of diacritics has been developed exclusively (or almost exclusively) for use with the letter 'alif (see Table 2) -- a special series of diacritics which is separate from the general system of diacritics denoting short vowels and syllable structure. However, even these special diacritics shown in Table 2 unfortunately do not completely resolve all the ambiguities connected with the letter 'alif without the presence of the rarely-used general vowel diacritics -- and the special diacritics of Table 2 are themselves quite often omitted in practice when writing Arabic.

Table 2

(For a partial explanation of these particular diacritics, see the listing in Appendix 1 below.)

Other ambiguities of current Arabic-script orthography will be referred to below, in the discussion of how these problems are resolved in the Tengwar orthography.

Listing of Tengwar characters to write Arabic

The following three diagrams show the use of Tengwar "primary" letters and "additional" letters (as defined by Tolkien) -- together with other modified letters and diacritics -- to write the Arabic language. After each Tengwar character is placed first the corresponding symbol of Arabic-script orthography, and then a transcription of the relevant sound (using conventional phonetic and/or Semitological symbols which are often used in linguistic works, but which will not be explained here). If a Tengwar character corresponds to a sound which is not used in the basic vocabulary of standard written Arabic (but is used in Arabic dialects and/or the pronunciation of foreign words borrowed into Arabic), then the corresponding Arabic orthography symbol and "broad phonetic" transcription are shown in blue.

Note that some of the letters in the second diagram below were devised by myself for this Arabic Tengwar mode, and will not be found in standard Tengwa fonts available over the Internet. Instead, special fonts with the Arabic-specific Tengwar symbols are available at the download section of this page. The three diagrams below use the "Tengwar Quenya" font; versions of these diagrams with two other Tengwa fonts are included in the PDF version of this page.

The "primary" Tengwar letters in the first table above are arranged in a matrix where the first column represents dental/interdental sounds, the second column represents labial sounds, the third column represents velar sounds, and the fourth column represents uvular/pharyngeal/glottal sounds. (These values of the third and fourth columns are not inherent, but are due to an implementation choice for this one particular "mode" of using Tengwar.) The first two rows write stops (voiceless and voiced), the next two rows write non-sibilant fricatives (voiceless and voiced), the fifth row writes nasals, and the last row writes relatively "weak" sounds of various types.

Here the Tengwar letter for the [g] sound can be used in transcribing Egyptian dialect forms, but would also be used for transcribing the [g] sound contained in foreign loanwords into standard written Arabic. For example, the Arabic word meaning "English" is inconsistently written both إنجليزي (Egyptian or Egyptian-influenced form) and إنكليزي (non-Egyptian form), but could be written consistently with the symbol for the [g] sound in the Arabic Tengwar mode: . Similarly, the Tengwar letters for the [v] and [p] sounds could be used for transcribing these sounds in foreign loanwords into Arabic -- while the Tengwar symbols for long and short [e] and [o] vowels provide an orthography for sounds which simply can't be written at all using the current Arabic-language orthography (even if full diacritics are used!), despite the fact that these sounds occur in Arabic dialect words and foreign loanwords into Arabic.

The Tengwar diacritic symbols indicate the presence of a vowel after the consonant-letter they are placed over (or a modification of the pronunciation of the consonant-letter they are placed under). In the diacritics table above, the position of letters with respect to the diacritics which modify them is shown by including dummy letters in light gray. Short vowels and consonant-doubling are not transcribed in Tengwar Arabic mode unless optional diacritics are used (as with current Arabic orthography).

The symbols for "emphatic" letters in the second table above were devised by merging the Tengwar letter for the phonological feature of "emphasis" or 'itbaq (useful in transcribing Arabic dialects) with other Tengwar letters. The letters chosen for this purpose may not always be obvious from the point of view of current standard pronunciation, but they were selected partly to reflect historical pronunciation, and partly to avoid creating any character with three "bows". Also, the symbol for an ordinary long [a] vowel is a modification of Tolkien's Tengwar "long carrier" character, and the letters for the other standard long vowels were created by merging this symbol with other letters. Finally, this Tengwar Arabic mode retains one use of consonant diacritic dots -- two dots under a letter indicates palatalization (this diacritic can also be used to create letters for additional alveopalatal/palatal sounds, useful in writing colloquial Arabic dialects, as discussed in Appendix 2 below).

Short Examples

The table below gives some brief Arabic text snippets which illustrate the usage of the Tengwar symbols given above. Each phrase is given first in conventional Arabic orthography, then in three different Tengwa fonts -- a basic font (Tengwar Quenya), a "calligraphic" font (Tengwar Elfica), and a "sans-serif" font (Tengwar Elfetica) -- then in phonemic/phonetic transcription, and finally in English translation:

A Tengwar transcription of a connected Arabic text is given on this page below.

Explanation of Details

Here are some notes on the details of writing Arabic in this Tengwar "mode":

-- The Tengwar "short carrier" symbol (which resembles a dotless Latin letter "i") is used to represent the Arabic letter 'alif when it is hamzat-ul-wasl (i.e. marks a deletable short vowel at or near the beginning of a word), in masculine plural verb forms ending in a [w] sound or long [u] vowel sound, in most indefinite accusative singular triptote endings not preceded by ta' marbutah ( ة ) -- and in any other cases where 'alif of current Arabic orthography does not represent either a glottal stop consonant sound or a long [a] vowel (perhaps including the final 'alif of the first-person singular subject pronoun?):

Examples of the use of the "short carrier" symbol in Tengwar Arabic mode

-- Semivowels ([y] and [w] sounds) and long high vowels (long [i] and [u] sounds) are strictly distinguished in the Tengwar Arabic mode, so that that no single Tengwar symbol can write both a long [i] vowel and a following [y] simultaneously (as the letter ya' sometimes does in current Arabic-script orthography). Therefore, the nisbah adjective suffix must always be written with a sequence of long [i] Tengwa + [y] Tengwa (as shown in the case of the words `arabiyah and qadhdhafi above). This also has the effect of distinguishing certain pairs of forms which are not ordinarily distinguished in current Arabic orthography (unless the full panoply of diacritics is used):

Examples of orthographic contrasts created by vowel/semivowel distinctions of Tengwar Arabic mode

-- Those words which in current Arabic-script orthography still retain remnants of the archaic tendency not to write long [a] sounds with the letter 'alif are regularized in the Arabic Tengwar mode:

Regular use of long [a] Tengwar symbol

-- The combination of the tengwar glottal-stop symbol with a short [i] vowel diacritic over it doesn't have any inherently special status in Tengwar Arabic mode, but it may be used with greater frequency than other consonant letter + short-vowel diacritic combinations if there is a desire to preserve the information signalled in the current orthography by 'alif with hamzah below (at or near the beginning of a word), or by ya' with hamzah next to a short [i] vowel (in the middle of a word):

Use of Tengwar short-i diacritic to preserve hamzah vowel information

-- Since hamzat-ul-wasl (a deletable short vowel at the beginning of a word's main stem, or the definite article) and hamzat-ul-qat` (a glottal-stop consonant sound at the beginning of a word's main stem) are written with completely different Tengwar letters, the contrast between them is strictly obligatory in Tengwar Arabic mode (as opposed to current Arabic orthography, where the contrast is indicated by the presence of a somewhat optional hamzah diacritic). The following table shows the contrast between a verbal noun or masdar belonging to the Stem IV or yuf`ilu verbal derivation (with true glottal stop) and a verbal noun belonging to the Stem VIII or yafta`ilu verbal derivation (with hamzat-ul-wasl):

Obligatory distinction between hamzat-ul-wasl and hamzat-ul-qat`

-- One goal of Tengwar Arabic mode is to make sure that the boundary between the main stem of a word and any attached syntactic clitic prefixes is always completely unambiguous. These syntactic prefixes are wa- "and", fa- "and", bi- "in, by", li- "to, for", ka- "like", la-/l- intensive sentence particle, sa- future tense marker, 'a- interrogative particle, and al- "the". In current Arabic orthography, attaching one (or more) of these prefixes to a word often changes the visual appearance of what it attaches to (due to the cursive contextual nature of Arabic script), and it is not always immediately obvious what is part of a syntactic clitic prefix, and what is part of the word itself (the main stem).

In Tengwar Arabic mode, the definite article al- "the" is always written with a special ligature symbol (combining the Tengwar "short carrier" symbol, which writes hamzat-ul-wasl, with the Tengwar letter for the [l] sound), as seen in various tables above. This means that the definite article is written unambiguously -- and since no other clitic prefix can come after the definite article, therefore whenever a word has a definite article, the boundary between clitic prefixes and the main stem always falls immediately after the definite article.

Similarly, any hamzat-ul-wasl which is not part of the definite article always occurs at the beginning of a word's stem, so that if the Tengwar "short carrier" symbol occurs near the beginning of an orthographic word (and is not part of the definite article ligature), then the boundary between the syntactic clitic prefixes and the main stem always falls immediately before the "short carrier" symbol.

For these reasons, the "short carrier" symbol is NEVER deleted when syntactic prefixes are added to a word -- unlike in current Arabic orthography, where the letter 'alif which marks hamzat-ul-wasl is sometimes deleted:

Retention of Tengwar "short carrier" symbol (hamzat-ul-wasl) after syntactic clitic prefixes

So if a word contains one or more syntactic clitic prefixes, but no definite article or hamzat-ul-wasl is present, then in Tengwar Arabic mode a small high hyphen is inserted between the last prefix and the main stem of the word -- as in the example form fa-l-takun shown above -- in order to make clear the location of the beginning of the main stem of the word (which otherwise might be ambiguous). For further examples of such hyphens, see the Tengwar transcription of a connected Arabic text below.

-- The order in which Tengwar characters are listed in the three large tables above might be suitable for arranging characters into an ordering based on formal origin (i.e. basic "primary" letters, basic "additional" letters, "primary" letters with added marks, "additional" letters with added marks, letters used to write vowels, miscellaneous other letters, and diacritics). However, for purposes of general sorting and alphabetization, an ordering much more like the following would probably be used:

Alphabetical order of Tengwar Arabic mode letters

In the table above, Tengwar letters connected by an equals sign "=" would be treated equivalently in sorting words into alphabetical order. The definite article ligature can be sorted as a sequence of its two component letters, but in most cases the definite article is simply ignored in sorting (since it is a syntactic prefix which is not really an integral part of the word itself -- with a few semi-dubious exceptions such as Allah and alladhani "those two who").
Note that the following Tengwar letters never occur at the beginning of a word: -- in addition to these two letters, which only occur at the end of a word: .

Tengwar transcription of connected Arabic passage

The following is an example of a connected Arabic text (Chapter 1 of the Qur'an, or surat-ul-fatihah) written in Tengwar Arabic mode:

The above text displays a version of Tengwar Arabic mode with all optional diacritics added, including word-final inflectional short vowels ('i`rab). There are fewer diacritics in this text than in a fully-diacriticized version of this same passage in current Arabic-script orthography, for several reasons: First, since long vowels are indicated completely unambiguously with basic Tengwar letters, Tengwar vowel diacritics are needed only to indicate short vowels and diphthongs. Second, the lack of any short-vowel diacritic over a Tengwar "short carrier" symbol or definite article ligature near the beginning of a word indicates that a deletable word-initial vowel has in fact been deleted (without any need for a special hamzat-ul-wasl diacritic). Third, the assimilation of the [l] consonant of the definite article to a following "sun letter" (coronal consonant) can be assumed, so that the doubling of the consonant after the definite article does not have to be explicitly marked. And fourth, there are almost no silent letters in Tengwar Arabic mode (except for the "short carrier" symbol, which is set aside for that purpose) -- so if full diacritics are being used, then any Tengwar letter which does not have a short vowel diacritic over it, is not directly followed by a long-vowel Tengwar , and is not itself a "short carrier" symbol or long-vowel Tengwar , will almost certainly represent a consonant sound which is not followed by a vowel sound. Therefore there is no real need for a special symbol to mark consonants not followed by vowels (such as the sukun diacritic of current Arabic orthography).

In general terms, one can say that all of the contrasts written by the special diacritics of Table 2 above are handled by the use of separate base letters in the Tengwar Arabic mode. Similarly, the Arabic-script entities ya' and 'alif maqsurah -- the contrast between which is marked only by the somewhat-optional presence of consonantal diacritic dots in current orthography (see Appendix 1) -- are written by distinct base letters in the Tengwar Arabic mode (e.g. على `ala "on" vs. علي -- either `alayya "on me" or `aliy "Ali" ).

Pruning and streamlining the apparatus of diacritics makes it likelier that some useful disambiguating diacritics will be used in ordinary text (i.e. outside of pedagogical works and sacred scriptures).

For an interlinear version of the above passage (with current Arabic-script orthography and Tengwar on alternate lines), view the PDF-file version of this web-page (see the "Downloads" section immediately below).

Downloads

Three non-HTML non-GIF files about the Tengwar Arabic mode can be obtained from this page, as well as the general "Tengwar Elfetica" font. These files are placed in the public-domain (with respect to what I have authored), and may be freely redistributed:

Here is basic information about the glyphs in the three Arabic Tengwar fonts contained in the large Zip file:

In addition, the "Tengwar-QuenyArabic" font contains larger and more distinct forms of vowel diacritics, for use with the characters of the TengwarQuenya and Tengwar-QuenyArabic fonts. Different widths of the circumflex (short-[a]) diacritic are mapped to keystrokes C, D, E, #; different widths of the overdot (short-[i]) diacritic are mapped to keystrokes B, G, T, %; and different widths of the left curl (short-[u]) diacritic are mapped to keystrokes M, J, U, &. Also, for the foreign/dialect vowels, different widths of the acute accent (short-[e]) diacritic are mapped to keystrokes V, F, R, $, and different widths of the right curl (short-[o]) diacritic are mapped to keystrokes N, H, Y, ^.

Links

• Specific font links:
  • Daniel Steven Smith created the "Tengwar Quenya" font (for which I have supplied matching Tengwar Arabic-mode letters in my "Tengwar-QuenyArabic" font), and his site is the home of the "Tengwar Quenya" font and his other fonts, and contains much general documentation on how to write languages with different Tengwar modes. You can also download the TengwarQuenya_v19E.zip font archive from a mirror site.
  • Ronald Kyrmse created the original version of the "Elfetica" font, and approved my revising it into the "TengwarElfetica" font -- with smaller relative x-height, remapping of characters, and a number of small detailed changes -- for which I have supplied matching Tengwar Arabic-mode letters in my "Tengwar-ElfeticArabic" font. His "Ondolindello" site contains a variety of information on Tengwar in English and Portuguese.
  • Enrique Mombello created the "Tengwar Elfica" font (for which I have supplied matching Tengwar Arabic-mode letters in my "Tengwar-ElficArabic" font). His website currently seems to be down, though there is a Spanish-language site under construction.
• General Tengwar / Tolkien linguistics sites:
  • The Tolkien Script Publishing contains some very detailed textbooks and source documentation on using Tengwar.
  • The Amanye Tenceli Tengwar information site.
  • The Ardalambion Tolkien linguistics site.
  • The Gwaith-i-Pheddain Tolkien linguistics site

Appendix 1 -- List of Tengwar corresponding to Arabic letters, in Arabic alphabetic order

Appendix 2 -- Use of Tengwar to write Arabic dialects

In addition to the [g] Tengwar and the [e] and [o] vowel diacritics discussed above, other symbols appropriate for transcribing Arabic colloquial dialect sounds can be derived by placing the two-dots diacritic under a velar or sibilant letter, or by placing the "emphasis" 'itbaq Tengwar letter before or over any other appropriate Tengwar letter. Here are some examples of Tengwar letters or combinations which could be used to transcribe Arabic colloquials:

Note that the Tengwar "emphasis" letter modifying the Tengwar letter for [l] is quite different from the Tengwar "emphasis" letter merged with the Tengwar letter for [l]. The former writes a velarized [l] sound, while the latter writes the letter dad.

Here's one hypothetical example of the use of Tengwar symbols to transcribe a colloquial Arabic dialect -- the following table shows a translation of the special symbols of Nizar Habash's "Palestinian Arabic Spelling Standardization Project" from Arabic script to Tengwar:

[For a brief paragraph about transcribing non-Arabic languages written with the Arabic alphabet into Tengwar, see the PDF version of this web-page, available at the download section of this page above.]

Go to a page on the most accurate form of the name of Jesus in Arabic.